MoE 技术
Mixture of experts (MoE) is a machine learning technique where multiple expert networks (learners) are used to divide a problem space into homogeneous regions. MoE represents a form of ensemble learning. They were also called committee machines.
1
衍生技术
7
相关产品
18
相关事件
相关专利
暂无数据
相关事件查看全部 (18)
Hyperbolic and Evidence-Prioritized Experts for Large Vision-Language Models
2026-06-02PRODUCT_LAUNCH影响: MEDIUM
PR2: Predictive Routing Replay for MoE-Based LLM Reinforcement Learning
2026-06-02PRODUCT_LAUNCH影响: MEDIUM
MoEIoU: Rethinking Bounding-Box Regression as a Mixture of Experts
2026-06-02PRODUCT_LAUNCH影响: MEDIUM
Beyond Task-Agnostic: Task-Aware Grouping for Communication-Efficient Multi-Task MoE Inference
2026-06-02PRODUCT_LAUNCH影响: MEDIUM
相关文章查看全部 (10)
When Model Merging Breaks Routing: Training-Free Calibration for MoE
ArXiv CS.CL2026-06-03
Rethinking the Role of Tensor Decompositions in Post-Training LLM Compression
ArXiv CS.AI2026-06-03
Expert-Aware Causal Tracing of Factual Recall in Sparse MoE Language Models
ArXiv CS.CL2026-06-03